Using an Enhanced LBG Algorithm to Reduce the Codebook Error in Vector Quantization

Authors:
Jörg Haber;Hans-Peter Seidel
Affiliations:
-;-
Venue:
CGI '00 Proceedings of the International Conference on Computer Graphics
Year:
2000

Citing 0
Cited 1

A novel approach for vector quantization using a neural network, mean shift, and principal component analysis-based seed re-initialization

Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we present a modification of the well-known LBG algorithm for the generation of codebooks in vector quantization. Our algorithm, denoted as the ILBG algorithm, reduces the codebook error of the LBG algorithm drastically in typical applications. In our experiments, we were able to achieve up to 75% reduction of the codebook error in only a few additional iteration steps. In the context of lossy image compression, this error reduction leads to an increase of 2-3dB of the peak-signal-to-noise-ratio (PSNR) in turn.